Long-term variations of the PM2.5 concentration identified by MODIS in the tropical rain forest, Southeast Asia

Abstract Aerosol and particulate matter are playing significant roles in the regional climate and environment in the tropical rain forest of Southeast Asia. Both satellite and ground observations showed significant seasonal variations in the PM 2.5 concentration and the aerosol optical properties during 2012–2014 in the Xishuangbanna tropical rain forest. The annual mean values of the PM 2.5 , aerosol optical depth (AOD), and the Angstrom exponent (α) were 34.3 ± 19.7 μg·m −3 , 0.54 ± 0.37, and 1.36 ± 0.20 in the dry season and 16.90 ± 5.08 μg·m −3 , 0.37 ± 0.10, and 1.07 ± 0.25 in the wet season, respectively. The results showed that 46.9% and 56.5% of the moderate-resolution imaging spectroradiometer (MODIS) C6 AOD data met the NASA accuracy requirements in the dry and wet season respectively and 17.1% and 17.7% of the seasonal mean systematically underestimated the ground-based data. There was a high correlation between PM 2.5 and AOD. The range of the correlation coefficient (R 2 ) was 0.69–0.85 in the dry season and 0.33–0.39 in the wet season. Linear regression functions of PM 2.5 and MODIS AOD were developed and used to retrieve the spatial and temporal distributions of the PM 2.5 in the tropical rain forest over the last decade (2006–2015). The annual mean PM 2.5 increased slightly in the region. The range of the PM 2.5 was 20–40 μg·m −3 in the wet season and 25–80 μg·m −3 in the dry season. In northern Thailand, northern Vietnam and the central district of Laos, PM 2.5 was up to the range of 50–80 μg·m −3 , which was mainly attributed to biomass burning in these areas.